A fully automatic method and device for leveling airborne electromagnetic data

Abstract

The application provides a kind of full-automatic leveling method and device for airborne electromagnetic data, first, mirror image period extension is carried out on each survey line data to avoid endpoint effect of data, and Grubbs criterion is used to remove abnormal data;Secondly, based on the extended data of survey area, the data of each survey line in the survey area is connected head to tail, and the variational mode decomposition is carried out on the data of the whole survey area;Finally, the first variational mode component after mean removal is taken as the leveling error, and the leveling error is removed from the original data as the leveling result.This method considers the characteristics of the leveling error along the survey line low-frequency variation, not only does not need to cut the line data, but also does not need to set parameters in the leveling process, and the whole leveling process can be fully automatic.

Description

Technical Field

[0001] This invention belongs to the field of aviation electromagnetic data calculation, and specifically relates to a fully automatic aviation electromagnetic data leveling method and device. Background Technology

[0002] Airborne electromagnetic surveying (AESS) is a geophysical exploration method based on Faraday's law of electromagnetic induction, using aircraft as the carrier. It boasts advantages such as high efficiency, wide coverage, and deep detection depth, making it suitable for mineral exploration and geological surveys in areas with complex topography and geology. However, AESS typically employs an "S"-shaped flight pattern. Inconsistencies in the flight status of different survey lines during the survey process can lead to strip-shaped leveling errors along the survey line direction, severely affecting the results. To eliminate these strip-shaped errors and improve the quality of the data, leveling processing of the airborne electromagnetic measurement data is necessary.

[0003] Chinese patent CN106970426A discloses "A method for leveling airborne electromagnetic data based on survey line differential and principal component analysis." This method first selects a baseline survey line and performs survey line differential on the survey area data along the survey line direction; based on the differential data of the survey area, principal component analysis is performed along the cutting line direction to obtain the differential form of the leveling error; finally, the leveling error of each survey line is iteratively calculated to obtain the leveling result. This method completes leveling by selecting a baseline survey line as the leveling standard, but it does not involve fully automatic leveling of airborne electromagnetic data.

[0004] Lin Zhenmin (1980) and Mao Yuxian (1995) proposed a pseudo-cut line leveling method for aeromagnetic data, which requires additional flight cut lines. Li Wenjie (2007) proposed a two-dimensional automatic leveling algorithm based on one-dimensional and two-dimensional filtering, combined with moving average filtering, which effectively removed leveling errors in frequency domain aeromagnetic data. Sun Dongming et al. (2010) and Chen Xiong et al. (2011) differentiated the pseudo-cut line data, determined its abrupt change points to identify the leveling area, and combined it with filters to achieve data leveling. Guo Tao et al. (2013) combined the total regional background field and the local regional background field, and used median filtering to achieve regional control leveling. This type of algorithm requires the use of filters and setting filter parameters. To date, a fully automatic leveling method for aeromagnetic data has not been found. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of the prior art by providing a fully automatic leveling method and apparatus for aviation electromagnetic data, which solves the technical problems of traditional leveling methods requiring cut line data assistance, filter parameter settings, or selection of leveling references.

[0006] To achieve the above objectives, a fully automated airborne electromagnetic data leveling method includes the following steps:

[0007] a. When entering airborne electromagnetic survey data, it is not necessary to grid the survey data in order to preserve the characteristics of the data itself;

[0008] b. Perform mirror periodic extension on the data of each survey line to remove the leveling error effect at the connection of the survey line data;

[0009] c. Use the Grubbs criterion to remove outlier data to avoid interference from the leveling process.

[0010] d. Perform variational mode decomposition on the preprocessed data. Based on the time-frequency characteristics of the leveling error in the measurement area and measurement line, introduce variational mode decomposition of adaptive signal processing. Search for the optimal solution of the intrinsic mode model through iterative calculation, determine the center frequency and bandwidth of each intrinsic mode component, realize adaptive subdivision of the signal frequency and effective separation of each component, thereby accurately extracting the low-frequency changing leveling error component.

[0011] e. Take the first variational mode component as the leveling error component, and take the first variational mode component after removing the mean as the leveling error.

[0012] f. Remove the leveling error of the mirror periodic extension part, and remove the corresponding leveling error from the original data as the leveling result;

[0013] g. Output the leveled data and plot it.

[0014] Step a describes the input of airborne electromagnetic survey data. To preserve the inherent characteristics of the data, it is not necessary to grid the survey data. For the airborne electromagnetic data of the survey area to be leveled, assuming there are l survey lines in the survey area, and each survey line contains n = [n1, n2, ..., nl] survey points, the survey data is represented as follows: There is no need to grid the data in the survey area.

[0015] Step b involves performing a mirror periodic extension on the data from each survey line to remove the leveling error effect at the data connection points. Because leveling errors exist in the survey area, the data levels of each survey line are inconsistent. If the data from each survey line in the survey area are connected end-to-end, a jump will occur at the connection points between the two survey lines. To avoid the endpoint effect during data decomposition, a mirror periodic extension is performed at both ends of each survey line. The number of extension measurement points at both ends of each survey line is set to... At this time, each survey line includes the following survey points:

[0016]

[0017] After mirror periodic extension of the data from each survey line, the survey area data is represented as follows:

[0018]

[0019] Step c describes the process of removing outlier data using the Grubbs criterion to prevent interference from outlier data during the leveling process. Outlier data in airborne electromagnetic data can affect the extraction of leveling errors. To achieve a fully automated leveling process, preprocessing of the survey area data is necessary to remove outlier data. Taking the first survey line as an example, the extended data for the first survey line is... When the residual between the measured point data and the mean of the measured line data satisfies Grubbs' criterion, it is considered that the measured point data will affect the extraction of the leveling error. The measured point data is derived from the median of the measured line. Data substitution:

[0020]

[0021] in, Let g(num,a) be the mean of the measurement data, and g(num,a) be the critical value coefficient, which depends on the number of measurements num and the significance level a. In this invention, g(num,a) is taken as 1.148. Let be the standard deviation of the survey line data. After removing outlier data from the first survey line in the survey area, it can be represented as:

[0022]

[0023] Following step c, abnormal data are removed from the data of each survey line in the survey area, and the data of each survey line in the survey area are then concatenated end to end. At this point, the preprocessed survey area data can be denoted as:

[0024] Step d describes performing variational mode decomposition on the preprocessed data. Variational mode decomposition searches for the optimal solution of the variational mode model through iterative calculation, determines the center frequency and bandwidth of each intrinsic mode component, thereby achieving adaptive frequency partitioning and effective separation of each component of the signal.

[0025] In order to extract the leveling error of low-frequency variations along the survey line in the survey area, the present invention sets the survey area data D. ext_gru After variational mode decomposition, two modal components u1(t) and u2(t) are obtained. To ensure good convergence of the variational mode decomposition, a quadratic penalty term and Lagrange multipliers are introduced. The specific steps are as follows:

[0026] (1) Variational mode decomposition parameter settings: Let the extended survey area data D ext_gru For the data to be decomposed, the quadratic penalty factor α is set to 10 times the length of the data to be decomposed, and the convergence criterion τ is set to 10. -6 .

[0027] Initialize the variational modal component u(t), the center frequency ω of the variational modal component, and the Lagrange multiplier λ, i.e., let:

[0028]

[0029] (2) Based on the set variational mode decomposition parameters, the variational mode component u(t), the center frequency ω of the variational mode component, and the Lagrange multiplier λ are updated iteratively. In the m-th iteration, the variational mode component is calculated using the following formula:

[0030]

[0031] Where δ(t) is the Dirac function, and * denotes convolution. According to Passevar's theorem, the sum of the squares of a function is equal to the sum of the squares of its Fourier transforms. Transforming equation (4) to the spectral domain, we obtain the following equation:

[0032]

[0033] Where sgn is the sign function. Utilizing the Hermitian matrix symmetry of the actual signal, the two terms in equation (5) can be written as half-space integrals over non-negative frequencies, while letting ω = ω + ω1:

[0034]

[0035] According to equation (6), the variational mode components of the m-th iteration can be obtained:

[0036]

[0037] The formula for calculating the center frequency of variational modal components is as follows:

[0038]

[0039] Similarly, transforming equation (8) to the spectral domain yields the center frequency of the variational mode component in the m-th iteration:

[0040]

[0041] The Lagrange multiplier update iterative formula is as follows:

[0042]

[0043] (3) Set the iteration termination condition for variational mode decomposition. When one of the terms in equation (11) is reached, the iteration terminates.

[0044]

[0045] (4) For variational mode components and Performing an inverse Fourier transform yields the time-domain decomposition results u1(t) and u2(t), thus completing the data D for the survey area.ext_gru The variational mode decomposition process.

[0046] Step e describes taking the first variational mode component (without the mean) as the leveling error. In this method, the first variational mode component yields a low-frequency leveling error component. This component is then subjected to mean-removal processing to obtain the preprocessed leveling error for the survey area.

[0047]

[0048] Step f involves removing the leveling error from the extended portion and removing the leveling error from the original data to obtain the leveling result. The leveling error D obtained in step e is... ext_error Data D is the original data after mirror periodic extension. ext_gru The corresponding leveling error is the leveling error obtained by calculating the original survey area data D, from D. ext_error Remove the extended portion:

[0049]

[0050] The leveling result of this invention is the leveling error corresponding to this part removed from the original data of the survey area (Equation (13)): D result =DD error .

[0051] Step g involves outputting the leveling results and plotting them. The output leveling results for the survey area are shown in Figure D. result And a map is generated based on the coordinates of each measuring point.

[0052] Compared with existing technologies, the beneficial effects of this invention are as follows: The fully automatic airborne electromagnetic data leveling method disclosed in this invention can quickly and effectively remove strip-shaped leveling errors in survey area data, such as... Figure 5 As shown; this invention is applicable to both aeromagnetic and electromagnetic data; this invention does not require auxiliary correction using survey area cutting lines, saving significant costs; the parameters used in this invention are all fixed and require no further adjustment, making it a completely automatic leveling method. Furthermore, from Figure 2 and Figure 4 As can be seen from the data, because this invention combines the characteristics of leveling error in the time domain and frequency domain, the characteristics of the data itself are preserved in the leveling result, and the leveling error is removed relatively completely. Attached Figure Description

[0053] Figure 1 A flowchart of a fully automated airborne electromagnetic data leveling method;

[0054] Figure 2 A plan view of the survey area for airborne magnetic data;

[0055] Figure 3 A graph showing the data curves for the airborne magnetic survey area;

[0056] Figure 4 A curve diagram of preprocessed airborne magnetic data for the survey area;

[0057] Figure 5 A plan view showing the leveling results of the airborne magnetic data survey area;

[0058] Figure 6 This is a plan view showing the leveling error of the airborne magnetic data measurement area. Detailed Implementation

[0059] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0060] It should be noted that the terms "first terminal," "second terminal," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects. The terms "first" and "second" refer to the technical solutions disclosed in detail in this invention and are not necessarily performed sequentially. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0061] The following is a further detailed description with reference to the accompanying drawings and embodiments:

[0062] This invention discloses a fully automated airborne electromagnetic data leveling method based on variational mode decomposition. Taking a set of airborne magnetic data detected by Geotech as an example, the survey area contains 117 survey lines (denoted as L10160-L11320), with a survey line spacing of 200m. The minimum number of survey points in the survey area is 569, and the maximum number of survey points is 733.

[0063] a. Inputting airborne electromagnetic survey data. For the airborne electromagnetic data of the survey area to be leveled, the survey area has 117 survey lines, and each survey line contains survey points n = [687,733,…,614,606]. The survey area data is represented as follows: like Figure 2 As shown, there is no need to grid the data in the survey area.

[0064] b. Perform mirror periodic extension on the data of each survey line. Due to leveling errors in the survey area, the data levels of each survey line are inconsistent. If the data of each survey line in the survey area are connected end to end, a jump will occur at the junction of the two survey lines. To avoid the endpoint effect during data decomposition, mirror periodic extension is performed at both ends of the survey line. The number of extension measurement points at both ends of each survey line is set to n. add = [343,366,…,307,303], where each survey line contains n survey points. ext =[1373,1465,…,1228,1212]. After mirror periodic extension of the data from each survey line, the survey area data is represented as follows:

[0065]

[0066] c. Use the Grubbs criterion to remove outlier data. Outlier data in airborne electromagnetic data can affect the extraction of leveling errors. To achieve a fully automated leveling process, preprocessing of the survey area data is necessary to remove outlier data. Taking the first survey line as an example, the extended data for the first survey line is... When the residual between the measured point data and the mean of the measured line data satisfies Grubbs' criterion, it is considered that the measured point data will affect the extraction of the leveling error. The measured point data is derived from the median of the measured line. Data substitution:

[0067]

[0068] in, This is the mean of the survey data. Let be the standard deviation of the survey line data. After removing outlier data from the first survey line in the survey area, it can be represented as:

[0069]

[0070] Following step c, abnormal data are removed from the data of each survey line in the survey area, and the data of each survey line in the survey area are then concatenated end to end. At this point, the preprocessed survey area data can be denoted as: like Figure 4 As shown.

[0071] d. Perform variational mode decomposition on the preprocessed data. Variational mode decomposition searches for the optimal solution of the variational mode model through iterative calculation, determines the center frequency and bandwidth of each intrinsic mode component, thereby achieving adaptive frequency partitioning and effective separation of each component of the signal.

[0072] In order to extract the leveling error of low-frequency variations along the survey line in the survey area, the present invention sets the survey area data D. ext_gru After variational mode decomposition, two modal components u1(t) and u2(t) are obtained. To ensure good convergence of the variational mode decomposition, a quadratic penalty term and Lagrange multipliers are introduced. The specific steps are as follows:

[0073] (1) Variational mode decomposition parameter settings: Let the extended survey area data D ext_gru For the data to be decomposed, the quadratic penalty factor α is set to 10 times the length of the data to be decomposed, and the convergence criterion τ is set to 10. -6 .

[0074] Initialize the variational modal component u(t), the center frequency ω of the variational modal component, and the Lagrange multiplier λ, i.e., let:

[0075]

[0076] (2) Based on the set variational mode decomposition parameters, the variational mode component u(t), the center frequency ω of the variational mode component, and the Lagrange multiplier λ are updated iteratively. In the m-th iteration, the variational mode component is calculated using the following formula:

[0077]

[0078] Where δ(t) is the Dirac function, and * denotes convolution. According to Passevar's theorem, the sum of the squares of a function is equal to the sum of the squares of its Fourier transforms. Transforming equation (4) to the spectral domain, we obtain the following equation:

[0079]

[0080] Where sgn is the sign function. Utilizing the Hermitian matrix symmetry of the actual signal, the two terms in equation (5) can be written as half-space integrals over non-negative frequencies, while letting ω = ω + ω1:

[0081]

[0082] According to equation (6), the variational mode components of the m-th iteration can be obtained:

[0083]

[0084] The formula for calculating the center frequency of variational modal components is as follows:

[0085]

[0086] Similarly, transforming equation (8) to the spectral domain yields the center frequency of the variational mode component in the m-th iteration:

[0087]

[0088] The Lagrange multiplier update iterative formula is as follows:

[0089]

[0090] (3) Set the iteration termination condition for variational mode decomposition. When one of the terms in equation (11) is reached, the iteration terminates.

[0091]

[0092] (4) For variational mode components and Performing an inverse Fourier transform yields the time-domain decomposition results u1(t) and u2(t), thus completing the data D for the survey area. ext_gru The variational mode decomposition process.

[0093] e. Take the first variational mode component after removing the mean as the leveling error. In this method, the first variational mode component yields a low-frequency leveling error component. This component is then subjected to mean removal processing to obtain the pre-processed leveling error for the survey area.

[0094]

[0095] f. Remove the leveling error from the extended portion, and use the leveling error removed from the original data as the leveling result. The leveling error D obtained in step e is... ext_error Data D is the original data after mirror periodic extension. ext_gru The corresponding leveling error is the leveling error obtained by calculating the original survey area data D, from D. ext_error Remove the extended portion:

[0096]

[0097] The leveling result of this invention is the leveling error corresponding to this part removed from the original data of the survey area (Equation (13)): D result =DD error .

[0098] g. Output the leveling results and plot them. Output the leveling results of the survey area (D). result And a map is generated based on the coordinates of each measuring point, such as Figure 5 As shown, the corresponding leveling error D error like Figure 6 As shown.

[0099] Example 2

[0100] A fully automatic aircraft electromagnetic data leveling device is provided.

[0101] The processor is suitable for implementing a fully automated airborne electromagnetic data leveling method as described above.

[0102] The processor is an integrated PC-specific processor or a mobile phone or tablet-specific processor.

[0103] as well as

[0104] The storage unit is adapted to store a fully automated airborne electromagnetic data leveling method as described above, wherein the fully automated airborne electromagnetic data leveling method as described above is loaded and executed by a processor.

[0105] This method can be implemented in portable smart devices such as laptops, mobile phones, and tablets, or integrated into certain specialized equipment and vehicles, such as specialized surveying instruments, aircraft, and aerial photography equipment.

[0106] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0107] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0108] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0109] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0110] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A fully automatic method for leveling airborne electromagnetic data, characterized in that: Includes the following steps, 1) Conduct aerial surveys of the area to be surveyed and record the aerial survey data D. The survey data obtained from the aerial survey includes data from multiple survey lines; 2) Enter the aerial survey data, remove abnormal areas from the data, and record the resulting survey area data as follows: 3). The aforementioned D ext_gru After variational mode decomposition, two modal components u1(t) and u2(t) will be obtained. 4) Solve for the leveling error D ext_error Substituting the modal component u1(t) into the following equation The leveling error D is obtained. ext_error 5) Solve for the leveling result D result =DD error The results of leveling the aerial survey data after leveling the data of each survey area were obtained. Output the leveling results of the survey area D result 6) Based on the leveling aerial survey data output in step 5), D result The graph was plotted to obtain the leveled aero-electromagnetic data map.

2. The fully automatic airborne electromagnetic data leveling method according to claim 1, characterized in that: Step 2), the removal of abnormal area data from the aerial survey data, also includes: The data at both ends of the survey line are mirrored periodically extended, and the Grubbs criterion is used to remove outlier data. During the data extension process, the number of extension measurement points at both ends of each measurement line is set as follows: At this time, each survey line includes the following survey points: After mirror periodic extension of the data from each survey line, the survey area data is represented as follows:

3. The fully automatic airborne electromagnetic data leveling method according to claim 2, characterized in that: During data extension, the residuals between the measured point data and the mean of the measured line data are calculated using the Grubbs criterion. When the residuals between the measured point data and the mean of the measured line data satisfy the condition, the measured point data is recorded as outlier and removed, and the median of the measured line is used as the criterion. Data substitution: Let g(num, a) be the mean of the measurement data, and g(num, a) be the critical value coefficient, which depends on the number of measurements num and the significance level a. The standard deviation of the survey line data; After removing the abnormal measurement point data from the survey line data, it is represented as follows:

4. The fully automatic airborne electromagnetic data leveling method according to claim 3, characterized in that: Then, abnormal data in each survey line of the survey area were removed sequentially. The survey area data after the removal process was recorded as follows:

5. The fully automatic airborne electromagnetic data leveling method according to claim 1, characterized in that: The D mentioned in step 3) ext_gru After variational mode decomposition, it also includes ① Substituting the variational modal component u(t), the center frequency ω of the variational modal component, and the Lagrange multiplier λ into the variational modal component calculation formula and iterating cyclically, we obtain the following at the m-th iteration: Where δ(t) is the Dirac function, and * denotes convolution; ② By transforming to the spectral domain and integrating, the variational mode components of the m-th iteration can be obtained: ③The formula for calculating the center frequency of the variational modal component is as follows: ④ Transform to the spectral domain to obtain the center frequency of the variational mode component in the m-th iteration. ⑤ The Lagrange multiplier update iterative formula is as follows: ⑥ Set an iteration termination condition; the iteration terminates when the termination condition is met. For the obtained variational mode components and Performing an inverse Fourier transform yields the time-domain decomposition results u1(t) and u2(t), thus completing the data D for the survey area. ext_gru Variational mode decomposition.

6. The fully automatic airborne electromagnetic data leveling method according to claim 5, characterized in that: The transformation to the spectral domain specifically refers to, Transform the equation obtained in step ① into the spectral domain form: sgn is a symbolic function.

7. The fully automatic airborne electromagnetic data leveling method according to claim 6, characterized in that: The integral is constructed by the Hermitian matrix, and the two terms of the spectral domain expression are written as half-space integrals over non-negative frequencies. Simultaneously, letting ω = ω + ω1, we obtain:

8. The fully automatic airborne electromagnetic data leveling method according to claim 3, characterized in that: Step 5) also includes Calculate the leveling error corresponding to the original survey area data D, starting from D. ext_error Remove the extended portion: Then, the leveling error D corresponding to the calculated portion is removed from the survey area data. error ; D result =D-D error 。 9. A fully automatic aviation electromagnetic data leveling device, characterized in that: include The processor is adapted to implement any one of the fully automatic airborne electromagnetic data leveling methods of claims 1-8; as well as The storage unit is adapted to store the fully automatic airborne electromagnetic data leveling method according to any one of claims 1-8, wherein the fully automatic airborne electromagnetic data leveling method according to any one of claims 1-8 is loaded and executed by a processor.

Images